Thanks for the background. That helps a whole bunch.
And I wonder who would want to *pay* for such certification of an open source product. Actually, with correct lobbying to health care organizations it might be possible to get them to collectively "foot the bill."
Richard Schilling
Web Integration Programmer
Affiliated Health Services
Mount, Vernon WA USA
http://www.affiliatedhealth.org
-----Original Message-----
From: Michael D. Weisner
To: [EMAIL PROTECTED]
Sent: 30/01/02 0:52
Subject: RE: New FDA Software Regulations 1/11/02
This is a very interesting topic - one that has been discussed at length
with FDA officials. Technically speaking, the FDA rules require a
submission (a 510k, pma, etc) prior to marketing. Now, the question has
always been, what constitutes marketing? To most of us, and the
Congress
that began this insane system in '76, marketing was synonymous with
selling
(for $money$). In 1976, the "medical market" was a very lucrative one,
to
say the least.
I have posed the question to several FDA personnel (at trade shows, like
Medical Manufacturing East in June) that if the medical device
(software) is
not sold, does it require "pre-marketing approval"? The response has
always
been "Of course it does!", but I wonder if that is so. I don't believe
that
it has ever been put to a test. It would seem that open source
software,
which is not offered for sale, would not be subject to these
regulations,
although, it may be seen as "promoted" via our efforts, which may be
viewed
as a form of marketing.
But, back to the reason for the FDA regulations. In 1976, the medical
industry was unregulated and the product quality, particularly software,
was
not good. The requirement that medical software be tested, verified and
validated is not really a bad thing. It is terrible to see a buggy
piece of
software anywhere, particularly in medicine. I think that open source
permits many persons to be involved in the verification process,
although we
must somehow track (and regulate) the releases so that the software
maintains a standard of high quality. I think that the problem is not
the
requirement of a quality control and assurance system but rather the
formalization of the FDA submission.
BTW, if we wanted to submit a 510k application to FDA, how would the
legal
portion be handled? If the product is neither owned or sold, who would
be
responsible for filing the documentation? In addition to FDA, the EU
has
similar regulations forming - how would they interpret open source
medical
software?
All thoughts welcome,
Mike
----- Original Message -----
From: "Neufeld, Jonathan" < [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]> >
To: < [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]> >
Sent: Tuesday, January 29, 2002 7:11 PM
Subject: RE: New FDA Software Regulations 1/11/02
> The document seems to assume a software application "market" that
doesn't
> always map too well onto open source. How do these guidelines apply
to
> systems that are not "marketed" per se? For example, those built
in-house
> (out of OS components) and strait open source + support type systems.
Or
is
> FDA approval only something that license sellers need to worry about?
>
> Jonathan
>
> > If one refers to the document referred to in the quote below:
> >
> > http://www.fda.gov/cber/gdlns/esigvalid.pdf
<http://www.fda.gov/cber/gdlns/esigvalid.pdf>
> >
> > one comes away with the impression that one has tiptoed past the
> > graveyard. The validation requirements do *not* seem to apply to
> > electronic patient records, *except* when those records are used in
> > federally funded research.
> >
> > Comments?
> >
>
